| Autor: |
Kolla, Om Vitesh, Kolla, Om Vivek, Nandini, D. Usha, Mary, S. Prince, Selvan, Mercy Paul |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3075 Issue 1, p1-10, 10p |
| Abstrakt: |
Identifying and analyzing elements in the video can be difficult due to changes in lighting, the appearance of the subject, and comparable off-target elements in the background. In this study, we used YOLOv5 to extract and classify objects in the video. A dataset called "Detection" was created using Python approaches for detecting objects. The detection dataset contains many categories of "x" photographs. The YOLOv5 models were adjusted and refined to identify real-time items such as humans, dogs, rifles, etc. An annotated Detection dataset was used to train YOLOv5, which has been fine-tuned for faster performance and improved detection accuracy. The created model was used to determine bounding boxes for the objects in the video. Additionally, each object is identified by its name and highlighted with a different color, making object recognition easier. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
